metal–ligand coordination graph neural networks predicting organometallic bonding, integrated with high-throughput structure generation datasets for open-shell chemistry the largest, most diverse dataset of quantum chemical properties on experimentally-synthesized transition metal complexes models across the periodic table property prediction, generative models, SOTA architectures, and uncertainty quantification built to handle the entire periodic table fingerprints from spectral graph theory physics-based molecular representations that encode 3D structure with symmetry enforced by construction, at a fraction of the cost of learned models
